The proposed research in sensory coding is a necessary first step in a rigorous approach to understanding how the visual signals that encode real visual scenes are transformed into the physiological correlate of perception or awareness. The proposed research aims to: 1) develop a novel experimental method for decoding and interpreting neural activity in individual channels of the primate lateral geniculate nucleus (LGN); 2) evaluate a simple nonlinear model of precortical visual processing that explains the results obtained in (1) using just the principles of monotonicity and differentiation, two common properties of precortical neurons in many sensory systems including the visual system. The proposed experiments will be performed in rhesus monkeys, a practical model for human perception. The monkey model will also expedite the use of experimental results in the design of visual-neural prosthetic devices aimed at restoring sight in humans.